Adenosine-Stress Dynamic Myocardial CT Perfusion Imaging Initial Clinical Experience
Total Page:16
File Type:pdf, Size:1020Kb
Color Figure(s): F1-3 Art: RLI200544/4 3:36 10/2/4 1؍balt6/z7l-ir/z7l-ir/z7l00610/z7l3090-10z xppws S ORIGINAL ARTICLE AQ:1 Adenosine-Stress Dynamic Myocardial CT Perfusion Imaging Initial Clinical Experience Gorka Bastarrika, MD, PhD,*† Luis Ramos-Duran, MD,* Michael A. Rosenblum, MD,‡ Doo Kyoung Kang, MD,*§ Garrett W. Rowe, BS,* and U. Joseph Schoepf, MD*‡ Single photon emission computed tomography1 is the most widely Objective: To evaluate the feasibility of adenosine-stress dynamic myocardial used modality, although cardiac magnetic resonance imaging volume perfusion imaging with dual source computed tomography (CT) for the (MRI)2 has demonstrated its superiority for detecting nontransmural qualitative and quantitative assessment of myocardial blood flow (MBF) com- perfusion defects mainly because ofits higher spatial resolution.3 pared with stress perfusion and viability magnetic resonance imaging (MRI). Ϯ The usefulness of multi detector-row CT (MDCT) for ruling Material and Methods: Ten patients (8 male, 2 female, mean age 62.7 out significant coronary artery stenosis4–6 and for providing prog- 7.1 years) underwent stress/rest perfusion and delayed-enhancement MRI, nostic information in patients with suspected coronary artery dis- and a cardiac CT protocol comprising prospectively electrocardiogram -trig- ease7–10 has repeatedly been demonstrated. Moreover, recent liter- gered coronary CT angiography, dynamic adenosine-stress myocardial per- ature suggests the feasibility of using MDCT as a standalone fusion imaging using a “shuttle” mode, and delayed enhancement acquisi- technology for integrative evaluation of coronary heart disease.11–15 tions. Two independent observers visually assessed myocardial perfusion The standard spiral acquisition mode of MDCT, however, cannot defects. For semi-quantitative evaluation, CT- and MRI-derived myocardial- dynamically evaluate the time-resolved passage of contrast medium to-left ventricular upslope indices were compared. Additionally, absolute through the myocardium and thus provides only limited information MBF was quantified based on dynamic perfusion CT and correlated with on the hemodynamic consequences of coronary artery disease. semi quantitative CT measurements. Myocardial perfusion analysis was The greater detector coverage of a recently introduced dual- performed on a segmental basis. Analysis used paired t tests, Wilcoxon source CT (DSCT) system16,17 may enable the performance of signed-rank test, linear correlation, and Bland-Altman statistics. dynamic myocardial volume perfusion imaging of the heart by Results: A total of 149 segments (93.1%) were suitable for analysis. means of a dedicated “shuttle” mode, comprising rapid electrocar- Sensitivity, specificity, positive and negative predictive values for detection diogram (ECG)-triggered image acquisition at 2 alternating table of myocardial perfusion defects at CT compared with MRI were 86.1%, positions during contrast medium infusion. This investigation aimed 98.2%, 93.9%, and 95.7%, respectively. Semiquantitative analysis of CT at determining the feasibility of applying this technique for the data showed significant differences between ischemic and nonischemic qualitative and (semi) quantitative CT assessment of myocardial myocardium with a signal intensity upslope that was comparable with Ϯ Ϯ Ͼ perfusion during adenosine stress using stress/rest perfusion and MRI-derived values (CT: 5.2 2 SI/s, MRI: 4.8 2.3 SI/s, P 0.05). delayed enhancement MRI as the reference standard. Moderate correlation was observed between absolute CT quantification of MBF and semi-quantitative CT measurements. Mean total dose length product for the entire cardiac CT protocol was 1290.4 Ϯ 233.3 mGy cm. MATERIALS AND METHODS Conclusion: Adenosine-stress volumetric first pass CT perfusion imaging is The study protocol was approved by our institutional review feasible and may enable the evaluation of qualitative and semi quantitative board and all patients gave written informed consent. Ten consec- parameters of myocardial perfusion in a comparable fashion as MRI. utive symptomatic subjects with known or high likelihood of coro- nary artery disease prospectively underwent stress/rest perfusion and Key Words: computed tomography, coronary vessels, angiography, delayed enhancement MRI and stress-perfusion and delayed en- myocardial perfusion hancement cardiac CT. Subjects with known contrast media allergy, (Invest Radiol 2010;45: 000–000) impaired renal function (creatinine Ͼ1.5 mg/dL), arrhythmia (eg, atrial fibrillation), claustrophobia, or MRI-incompatible implanted devices were excluded from participation. All subjects underwent both procedures within 24 hours. n clinical practice, the physiological significance of coronary Iartery stenosis is ordinarily assessed with myocardial perfusion Cardiac MRI Acquisition Protocol imaging modalities during pharmacologically induced hyperemia. MRI studies were performed using a 1.5-T system (Magne- tom Avanto, Siemens, Erlangen, Germany) using a 6-element phased array coil. Stress perfusion MRI was performed 3 minutes Received December 10, 2009; accepted for publication (after revision) February into the intravenous infusion of adenosine (140 g/kg/min Ad- 15, 2010. enoscan, Astellas, Tokyo, Japan) using steady-state free precession From the *Department of Radiology and Radiological Science, Medical Univer- sity of South Carolina, Charleston, SC; †Department of Radiology, University (SSFP, TrueFISP Siemens) perfusion sequences. Three short axis of Navarra, Pamplona, Spain; ‡Division of Cardiology, Department of Med- sections representative of basal, mid, and apical myocardial portions icine, Medical University of South Carolina, Charleston, SC; and §Depart- were acquired with the following parameters: repetition time (TR) ment of Radiology, Ajou University Hospital, Suwon, South Korea. 2.8 ms, echo time (TE) 1.21 ms, flip angle 50°, field of view 380 ϫ Supported by Bayer-Schering, Bracco, General Electric, Medrad, and Siemens (to ϫ U.J.S.). 80.2 mm, matrix 116 192, acquisition duration 150 ms per slice, G.B. is a consultant for General Electric, Medrad, and Siemens. U.J.S. is a slice thickness 10 mm, 50 measurements, and an acceleration factor consultant. of 2 (GRAPPA, Siemens). Ten minutes after stress perfusion, rest Reprints: U. Joseph Schoepf, MD, Department of Radiology and Radiological perfusion images were obtained in the same technique. Contrast Science, Medical University of South Carolina, Ashley River Tower, 25 Courtenay Dr, MSC 226, Charleston, SC 29401. E-mail: [email protected]. enhancement during stress and rest perfusion MRI was accom- Copyright © 2010 by Lippincott Williams & Wilkins plished with gadopentetate dimeglumine (Magnevist, Bayer-Scher- ISSN: 0020-9996/10/4506-0001 ing, Berlin, Germany; 0.05 mmol/kg each for rest and stress MRI Investigative Radiology • Volume 45, Number 6, June 2010 www.investigativeradiology.com | 1 Color Figure(s): F1-3 Art: RLI200544/4 3:36 10/2/4 1؍balt6/z7l-ir/z7l-ir/z7l00610/z7l3090-10z xppws S Bastarrika et al Investigative Radiology • Volume 45, Number 6, June 2010 perfusion imaging, 0.1 mmol/kg total), injected at 4 mL/s followed contrast medium enhanced with 50 mL of iopromide, followed by 50 by 15 mL of normal saline. In addition, functional analysis was mL of saline, injected at 6 mL/s. Including test bolus acquisition and performed with retrospectively ECG-gated 8 mm slice-thickness coronary CTA angiography, each patient thus received a total cine loops in short and long axis views using a SSFP sequence (TR: volume of 150 mL contrast medium and 135 mL saline. Studies 3.09 ms, TE: 1.3 ms, flip angle 80°, field of view 280 ϫ 375 mm, were obtained during end-inspiration with a standardized acquisition matrix 156 ϫ 192, 25 phases per cardiac cycle, no interslice gap, time of 30 seconds. If patients could not hold their breath for 30 in-plane resolution 1.7 ϫ 1.7 mm). Finally, delayed images were seconds, they were instructed to slowly release their breath and acquired using a phase sensitive inversion-recovery SSFP sequence continue breathing shallowly. Images were reconstructed with 3 mm (TR: 3.38 ms, TE: 1.4 ms, flip angle 45°, field of view: 340 ϫ 68.8 slice width every 2 mm with a medium sharpness convolution mm, matrix 127 ϫ 256, slice thickness 8 mm, no interslice gap). algorithm and then processed using the Volume Perfusion software (syngo VA31, Siemens). Cardiac CT Acquisition Protocol Finally, delayed enhancement studies were performed 6 min- All patients underwent cardiac CT using a DSCT system utes after perfusion imaging using a regular prospectively ECG- (SOMATOM Definition Flash, Siemens). Initially, single heart-beat triggered mode with image acquisition at 70% of the R-RЈ interval CT calcium scoring was acquired with the following parameters: at 80 kV and 320 mAs. For delayed enhancement imaging, no 2 ϫ 64 ϫ 0.6 mm detector collimation resulting in 2 ϫ 128 ϫ 0.6 functional image information was acquired. mm sections by means of the z-flying focal spot technique, 280 ms gantry rotation time, 120 kV tube potential, and 73 mAs per rotation Perfusion Data Analysis tube current time product. Subsequently prospectively ECG-trig- Two experienced radiologists independently evaluated MRI gered coronary CT angiography was performed. Contrast medium and CT studies blinded to clinical history. CT and MRI data were enhancement was achieved using a triphasic injection protocol with evaluated separately and in random order. Discordant findings were injection of